Spaces:
Sleeping
Sleeping
File size: 9,929 Bytes
19759e2 b697762 19759e2 b697762 19759e2 b697762 19759e2 b697762 19759e2 b697762 1cb6e78 b697762 1cb6e78 b697762 19759e2 b697762 19759e2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"The autoreload extension is already loaded. To reload it, use:\n",
" %reload_ext autoreload\n"
]
}
],
"source": [
"%load_ext autoreload\n",
"%autoreload 2"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"README.md \u001b[34mmodel_checkpoints\u001b[m\u001b[m recommender.py\n",
"app.py \u001b[34mnotebooks\u001b[m\u001b[m requirements.txt\n",
"\u001b[34mdata\u001b[m\u001b[m orchestrate_audio_data.py \u001b[34msrc\u001b[m\u001b[m\n"
]
}
],
"source": [
"!ls"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import librosa\n",
"import torch\n",
"from src import laion_clap\n",
"from glob import glob\n",
"import pandas as pd\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"model = laion_clap.CLAP_Module(enable_fusion=False, amodel= 'HTSAT-base')\n",
"model.load_ckpt(ckpt=\"music_audioset_epoch_15_esc_90.14.pt\")"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"def load_music_file(file_name):\n",
" audio_data, _ = librosa.load(file_name, sr=48000) # sample rate should be 48000\n",
" audio_data = audio_data.reshape(1, -1) # Make it (1,T) or (N,T)\n",
" # audio_data = torch.from_numpy(int16_to_float32(float32_to_int16(audio_data))).float() # quantize before send it in to the model\n",
" with torch.no_grad():\n",
" audio_embed = model.get_audio_embedding_from_data(x = audio_data, use_tensor=False)\n",
" return audio_embed\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"music_files = glob(\"/Users/berkayg/Codes/music-project/AudioCLIP/data/downloaded_tracks/*.wav\")"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"import pickle\n",
"with open(\"/Users/berkayg/Codes/music-project/laion-clap-project/curate-me-a-playlist/data/vectors/song_names.pkl\", \"rb\") as reader:\n",
" ls = pickle.load(reader)\n"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/var/folders/sr/r72219hj06x_1xvw7hhd517h0000gn/T/ipykernel_18860/3009710654.py:2: UserWarning: PySoundFile failed. Trying audioread instead.\n",
" audio_data, _ = librosa.load(file_name, sr=48000) # sample rate should be 48000\n",
"/Users/berkayg/miniforge3/envs/playlist-curator/lib/python3.10/site-packages/librosa/core/audio.py:183: FutureWarning: librosa.core.audio.__audioread_load\n",
"\tDeprecated as of librosa version 0.10.0.\n",
"\tIt will be removed in librosa version 1.0.\n",
" y, sr_native = __audioread_load(path, offset, duration, dtype)\n"
]
}
],
"source": [
"music_data = np.zeros((len(music_files), 512), dtype=np.float32)\n",
"for m in range(music_data.shape[0]):\n",
" music_data[m] = load_music_file(music_files[m])\n"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(1, 512)\n"
]
}
],
"source": [
"text_data = [\"This audio is a romantic song\"] \n",
"text_embed = model.get_text_embedding(text_data)\n",
"print(text_embed.shape)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"song_names = [k.split(\"/\")[-1] for k in music_files]"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"torch.Size([100, 1])\n"
]
}
],
"source": [
"with torch.no_grad():\n",
" ranking = torch.tensor(music_data) @ torch.tensor(text_embed).t()\n",
" ranking = ranking[:, 0].reshape(-1, 1)\n",
"print(ranking.shape)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>This audio is a romantic song</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Coldplay - Charlie Brown.wav</th>\n",
" <td>0.400684</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Sam Smith - I'm Not The Only One.wav</th>\n",
" <td>0.373561</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Pink Floyd - The Great Gig In The Sky - 2011 Remastered Version.wav</th>\n",
" <td>0.371584</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Christina Aguilera - You Lost Me.wav</th>\n",
" <td>0.370390</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Lana Del Rey - Yayo.wav</th>\n",
" <td>0.370379</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Queen - It's A Hard Life - Remastered 2011.wav</th>\n",
" <td>0.348699</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Teoman - Haziran.wav</th>\n",
" <td>0.331220</td>\n",
" </tr>\n",
" <tr>\n",
" <th>John Lennon - Imagine - Remastered 2010.wav</th>\n",
" <td>0.330397</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Sleeping At Last - Mars.wav</th>\n",
" <td>0.328770</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Adele - Someone Like You.wav</th>\n",
" <td>0.325650</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Coldplay - What If.wav</th>\n",
" <td>0.315717</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Adamlar - Orda Ortada.wav</th>\n",
" <td>0.306465</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Eric Clapton - Autumn Leaves.wav</th>\n",
" <td>0.305451</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Premiata Forneria Marconi - Impressioni di settembre.wav</th>\n",
" <td>0.295878</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Guthrie Govan - Lost in Rio.wav</th>\n",
" <td>0.284883</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" This audio is a romantic song\n",
"Coldplay - Charlie Brown.wav 0.400684\n",
"Sam Smith - I'm Not The Only One.wav 0.373561\n",
"Pink Floyd - The Great Gig In The Sky - 2011 Re... 0.371584\n",
"Christina Aguilera - You Lost Me.wav 0.370390\n",
"Lana Del Rey - Yayo.wav 0.370379\n",
"Queen - It's A Hard Life - Remastered 2011.wav 0.348699\n",
"Teoman - Haziran.wav 0.331220\n",
"John Lennon - Imagine - Remastered 2010.wav 0.330397\n",
"Sleeping At Last - Mars.wav 0.328770\n",
"Adele - Someone Like You.wav 0.325650\n",
"Coldplay - What If.wav 0.315717\n",
"Adamlar - Orda Ortada.wav 0.306465\n",
"Eric Clapton - Autumn Leaves.wav 0.305451\n",
"Premiata Forneria Marconi - Impressioni di sett... 0.295878\n",
"Guthrie Govan - Lost in Rio.wav 0.284883"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"pd.DataFrame(ranking, columns=[text_data[0]], index=song_names).nlargest(15, text_data[0])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "playlist-curator",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|